CN114513838A - Moving edge calculation method, frame, and medium - Google Patents
Moving edge calculation method, frame, and medium Download PDFInfo
- Publication number
- CN114513838A CN114513838A CN202210188436.2A CN202210188436A CN114513838A CN 114513838 A CN114513838 A CN 114513838A CN 202210188436 A CN202210188436 A CN 202210188436A CN 114513838 A CN114513838 A CN 114513838A
- Authority
- CN
- China
- Prior art keywords
- base station
- task
- information
- intelligent equipment
- macro base
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000004364 calculation method Methods 0.000 title claims abstract description 78
- 238000004891 communication Methods 0.000 claims abstract description 62
- 238000005265 energy consumption Methods 0.000 claims abstract description 38
- 238000000034 method Methods 0.000 claims abstract description 34
- 230000005540 biological transmission Effects 0.000 claims description 14
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 description 14
- 230000001413 cellular effect Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000010295 mobile communication Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
- H04W52/02—Power saving arrangements
- H04W52/0203—Power saving arrangements in the radio access network or backbone network of wireless communication networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
- H04W52/02—Power saving arrangements
- H04W52/0203—Power saving arrangements in the radio access network or backbone network of wireless communication networks
- H04W52/0206—Power saving arrangements in the radio access network or backbone network of wireless communication networks in access points, e.g. base stations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
- H04W52/04—TPC
- H04W52/18—TPC being performed according to specific parameters
- H04W52/26—TPC being performed according to specific parameters using transmission rate or quality of service QoS [Quality of Service]
- H04W52/265—TPC being performed according to specific parameters using transmission rate or quality of service QoS [Quality of Service] taking into account the quality of service QoS
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a moving edge calculation method, a frame and a medium. The method comprises the following steps: the macro base station integrates the task information to be processed, the first network communication state, the second network communication state, the first computing resource information and the second computing resource information by a task engine according to the received information, and constructs global task information, network communication state and computing resource information; delivering the geographic position information to an intelligent traffic engine, and predicting the stay time of the mobile intelligent equipment in the communication range of the macro base station by combining intelligent traffic information; starting a scheduling engine, and making a scheduling strategy of the current time slice by using the received information; and informing the intelligent equipment, the micro base station and the edge cloud server of the scheduling strategy through a network. Compared with the prior art, the embodiment of the invention enlarges the application range, improves the task distribution accuracy and can consider the energy consumption of the system more comprehensively.
Description
Technical Field
The present invention relates to the field of mobile communications technologies, and in particular, to a mobile edge calculation method, a mobile edge calculation framework, and a mobile edge calculation medium.
Background
The development of mobile communication technology has also shown a tendency to accelerate, as has the development of other modern technologies. The internet of things equipment such as smart cars and mobile phones can be accessed to the internet through a cellular network, a low-power-consumption wide area network and the like, and the state of the surrounding environment can be sensed by using the equipped sensors. For example, in the fields of car networking and augmented reality, these applications require real-time processing of the acquired video information and feedback of the results to the user. A large number of intensive computing tasks inevitably accelerate the energy consumption of the Internet of things equipment and shorten the service life of the Internet of things equipment. The combination of the internet of things equipment and the cloud is a main mode of the application of the internet of things. However, the long-distance communication between the internet of things device and the remote cloud platform has the problem of unstable network transmission delay, which will cause the delay of the application of the internet of things to be too long, and cannot meet the application with clear requirements on delay.
In order to solve the problem, two technical schemes are provided in the prior art, the first study is carried out that when a plurality of mobile intelligent devices unload tasks to a universal cloud server under the MIMO-based multi-cell base station scene, the energy consumption is minimum, and an iterative algorithm based on a continuous convex approximation technology is provided. The set scene is as follows: there are two cellular sites in the network, each cellular site can provide service for 6 active users, and it is assumed that there are 4 users in each cellular site that require task offload service, and 2 other users are only responsible for data transmission, and the performance of the analysis algorithm is analyzed by changing a series of constraint parameters. The second study has studied the problem of lowest task offloading energy consumption in a multi-device scenario. Considering a small wireless base station scene, 30 mobile intelligent devices are randomly distributed in a base station communication range, firstly, each mobile intelligent device only performs local calculation, then all the mobile intelligent devices unload tasks to a cloud server for execution, and finally, online task unloading scheduling is performed according to an optimization algorithm, and system performance is analyzed by comparing energy consumption under three conditions.
However, the number of devices and the number of base stations in the above technical solution are small, which is not enough to meet the requirements of a real application scenario of the mobile intelligent device, and the stay time and the participation calculation time of the device cannot be accurately predicted, which affects the accuracy of task distribution. In addition, the technical scheme only focuses on the optimal energy consumption of the mobile intelligent device in the task execution process, and does not relate to the energy consumption of the base station, the association between the mobile intelligent device and the base station and other operation conditions. Therefore, the prior art has the problems of limited application, poor task distribution accuracy and incomplete consideration of system energy consumption.
Disclosure of Invention
The invention provides a mobile edge calculation method, a mobile edge calculation framework and a mobile edge calculation medium, which are used for expanding the application range, improving the task distribution accuracy and considering the system energy consumption more comprehensively.
According to an aspect of the present invention, there is provided a moving edge calculation method, including:
the mobile intelligent equipment provides task information to be processed, a first network communication state, first computing resource information and geographical position information to the macro base station;
the micro base station provides a second network communication state to the macro base station;
the edge cloud server provides second computing resource information of the virtual machine to the macro base station;
the macro base station integrates the task information to be processed, the first network communication state, the second network communication state, the first computing resource information and the second computing resource information by a task engine according to the received information, and constructs global task information, network communication state and computing resource information;
the macro base station delivers the geographic position information to an intelligent traffic engine, and the residence time of the mobile intelligent equipment in the communication range of the macro base station is predicted by combining intelligent traffic information;
the macro base station starts a scheduling engine, and a scheduling strategy of the current time slice is formulated by using the received information;
and the macro base station informs the intelligent equipment, the micro base station and the edge cloud server of the scheduling strategy through a network.
Optionally, the scheduling policy includes: at least one of a device task offload scheduling policy, a user-base station associated scheduling policy, and a base station sleep scheduling policy.
Optionally, after the macro base station notifies the smart device, the micro base station, and the edge cloud server of the scheduling policy through a network, the method further includes:
the edge cloud server provides computing service for the corresponding intelligent equipment;
after the calculation is finished, the calculation result is uploaded to the internet for further analysis and processing according to the requirement, or the calculation result is returned to the intelligent equipment.
Optionally, after the macro base station notifies the smart device, the micro base station, and the edge cloud server of the scheduling policy through a network, the method further includes:
the intelligent equipment obtains base stations to be associated, the amount of local execution of tasks and the amount of execution of unloading to the edge cloud server; the base station comprises a macro base station and a plurality of micro base stations;
the base station determines whether the current time slice needs to be started and the intelligent device to be associated.
Optionally, the scheduling engine uses the received information to make a scheduling policy of the current time slice, including:
establishing a calculation formula of the local task surplus of the intelligent equipment;
establishing a task surplus calculation formula of the virtual machine in the edge cloud server in the time slice;
establishing a calculation formula of the energy consumption of the intelligent equipment;
establishing an energy consumption calculation formula of the base station; the base station comprises a macro base station and a plurality of micro base stations;
an objective function is established that minimizes the overall energy consumption of the smart device and the base station, subject to constraints.
Optionally, the constraint condition comprises at least one of the following conditions:
each intelligent device establishes association with at most one base station;
any base station can be associated with a preset number of intelligent devices in each time slice at most;
the intelligent device and the base station can establish association only when the uplink SINR of the intelligent device and the base station is larger than a target SINR;
in any time slice, the computing resources of the virtual machine distributed by the edge cloud server for the intelligent equipment do not exceed the physical resources of the edge cloud server;
within each time slice, the tasks processed locally by the intelligent equipment do not exceed the computing capacity of the intelligent equipment;
in each time slice, the task amount unloaded by the intelligent equipment does not exceed the transmission capability of the intelligent equipment;
in each time slice, the sum of the tasks executed locally by the intelligent equipment and the tasks executed by unloading does not exceed the current task queue of the intelligent equipment;
in each time slice, the sum of the task execution time and the transmission time of the intelligent equipment does not exceed the stay time of the intelligent equipment in the current base station communication range;
only when the base station is in an open state, the intelligent equipment can be associated with the base station;
the average delay of the task can meet the delay requirement of the application.
Optionally, the mobile edge calculation method is based on a 5G network architecture and a smart traffic system.
Optionally, the mobile smart device comprises: at least one of a smartphone and a smart car.
According to another aspect of the present invention, there is provided a moving edge calculation framework, comprising:
the mobile intelligent equipment is used for providing task information to be processed, a first network communication state, first computing resource information and geographical position information to the macro base station;
a micro base station for providing a second network communication state to the macro base station;
the edge cloud server is used for providing second computing resource information of the virtual machine for the macro base station;
the macro base station is used for integrating the task information to be processed, the first network communication state, the second network communication state, the first computing resource information and the second computing resource information by a task engine according to the received information to construct global task information, network communication state and computing resource information;
the macro base station delivers the geographic position information to an intelligent traffic engine, and the residence time of the mobile intelligent equipment in the communication range of the macro base station is predicted by combining intelligent traffic information;
the macro base station starts a scheduling engine, and a scheduling strategy of the current time slice is formulated by using the received information;
and the macro base station informs the intelligent equipment, the micro base station and the edge cloud server of the scheduling strategy through a network.
According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the moving edge calculation method according to any one of the embodiments of the present invention when the computer instructions are executed.
According to the embodiment of the invention, a macro base station is arranged to integrate task information to be processed, a first network communication state, a second network communication state, first computing resource information and second computing resource information by a task engine according to received information, and construct global task information, network communication state and computing resource information; the geographical position information is delivered to an intelligent traffic engine, and the stay time of the mobile intelligent equipment in the communication range of the current macro base station is predicted by combining the intelligent traffic information; and starting a scheduling engine, and making a scheduling strategy of the current time slice by using the received information. The macro base station selects the most appropriate base station for the mobile intelligent equipment to be associated according to the current system information, and the micro base station which is not associated with the mobile intelligent equipment is enabled to enter a sleep state, so that energy is saved. Compared with the prior art, the embodiment of the invention can predict the time of the mobile intelligent device participating in the calculation through the intelligent traffic system, perform task unloading scheduling, device-base station association and base station sleep scheduling, minimize the overall energy consumption of the mobile intelligent device and the base station, simultaneously ensure the service quality requirement (mainly a delay requirement) of the application, and can be widely applied to a 5G network mobile calculation scene. Therefore, the embodiment of the invention expands the application range of the mobile edge calculation, improves the task distribution accuracy and considers the energy consumption of the system more comprehensively.
It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present invention, nor do they necessarily limit the scope of the invention. Other features of the present invention will become apparent from the following description.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic structural diagram of a moving edge calculation framework according to an embodiment of the present invention;
fig. 2 is a schematic flow chart illustrating a moving edge calculation method according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating another moving edge calculation method according to an embodiment of the present invention;
fig. 4 is a schematic flowchart of a calculation method for making a scheduling policy by a scheduling engine according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The embodiment of the invention provides a mobile edge calculation method. The method can be based on a 5G network architecture and an intelligent traffic system. The method can be suitable for the Internet of things, helps the mobile intelligent device to complete complex calculation tasks (such as VR, AR and smart city), and simultaneously minimizes the overall energy consumption of all the mobile intelligent devices and base stations participating in calculation in the whole network range.
The mobile edge computing is to merge a cloud computing platform in a traditional centralized data center with a mobile network, sink network services to a wireless access network, sink computing capacity to a distributed base station and an edge cloud, add functions such as computing and storage at the edge of the wireless network, upgrade a traditional wireless base station to an intelligent base station, provide cloud computing capacity at a place close to an intelligent terminal, provide task unloading service for a user, and further improve network utilization efficiency.
The intelligent traffic system is a comprehensive traffic and transportation management system which is established by effectively integrating technologies such as cloud computing, internet of things, big data processing and artificial intelligence and is capable of playing a role in a large range in all directions, real-time, accurate and efficient mode. The traffic system has the capabilities of perception, interconnection, analysis, prediction, control and the like, fully guarantees traffic safety, exerts the efficiency of traffic infrastructure, and improves the operation efficiency and the management level of the traffic system. The embodiment of the invention can obtain the stay time of the mobile intelligent equipment in the range of the base station by using the intelligent traffic system.
The 5G network architecture has the important characteristics of network function virtualization, supporting Software Defined Networking (SDN), bringing cloud computing and cloud storage to the edge of the network, creating a carrier-class service environment with high performance, low delay and high bandwidth, and accelerating distribution and downloading of various contents, services and applications in the network.
To describe the technical solution provided by the embodiment of the present invention more clearly, the moving edge calculation framework provided by the embodiment of the present invention is introduced first and named ITMED. Fig. 1 is a schematic structural diagram of a moving edge calculation framework according to an embodiment of the present invention. Referring to fig. 1, the framework is mainly composed of a mobile smart device 210, a macro base station 221, a micro base station 222, and an edge cloud server 230. Assume that the mobile smart device 210 is represented by a set N ═ {1, 2., N }, and the base station 220 is represented by a set M ═ {1, 2., M }, where the set M of base stations includes 1 macro base station 221 and M-1 micro base stations 222. The framework considers the macro base station 221 as a central control center, and a task engine process, an intelligent traffic engine process and a scheduling engine process are set in the macro base station 221. The task engine process can collect information of tasks to be processed, network communication states and available computing resources of the mobile intelligent device, network communication states of the micro base stations 222 and computing resource information of the edge cloud server, and estimate execution time and energy consumption of current tasks in the system by using the information. The intelligent traffic engine process can collect the geographic position information of the mobile intelligent device and predict the stay time of the mobile intelligent device in the communication range of the current base station by combining the intelligent traffic information. And the scheduling engine process completes task unloading scheduling, equipment-base station association and base station sleep scheduling of the mobile intelligent equipment in the system according to the information collected and processed by the task engine process and the intelligent traffic engine process, and ensures that task transmission and calculation are completed within the stay time range of the mobile intelligent equipment.
Fig. 2 is a flowchart illustrating a moving edge calculation method according to an embodiment of the present invention. Referring to fig. 2, the moving edge calculation method includes the steps of:
s110, the mobile intelligent device provides to-be-processed task information, a first network communication state, first computing resource information and geographical position information to the macro base station.
The mobile smart device (hereinafter, may be simply referred to as a device) may be in various forms, for example, a smart phone, a smart car, and the like. The mobile intelligent device collects data by utilizing a sensor equipped with the mobile intelligent device, wherein the sensor can be a gyroscope, an acceleration sensor and the like. The ITMED framework operates by taking a time slice t as a unit, and in each time slice, the mobile intelligent equipment needs to provide self task information to be processed, a first network communication state, first computing resource information and geographical position information to the macro base station. In the ITMED framework, the mobile intelligent device can perform local calculation, and can also utilize a wireless network to access the base station, unload tasks to the edge cloud server under the scheduling of the macro base station, and fully utilize resources of the edge cloud server to perform calculation.
And S120, the micro base station provides a second network communication state for the macro base station.
S130, the edge cloud server provides second computing resource information of the virtual machine for the macro base station.
S140, the macro base station delivers the task information to be processed, the first network communication state, the second network communication state, the first computing resource information and the second computing resource information to a task engine for integration according to the received information, and constructs overall task information, network communication state and computing resource information.
S150, the macro base station delivers the geographic position information to the intelligent traffic engine, and the intelligent traffic engine is combined with the intelligent traffic information to predict the stay time of the mobile intelligent device in the communication range of the current macro base station.
S160, the macro base station starts a scheduling engine, and a scheduling strategy of the current time slice is formulated by using the received information.
Optionally, the scheduling policy includes: at least one of a device task offload scheduling policy, a user-base station associated scheduling policy, and a base station sleep scheduling policy.
S170, the macro base station informs the intelligent equipment, the micro base station and the edge cloud server of the scheduling strategy through the network.
Optionally, after the macro base station notifies the smart device, the micro base station, and the edge cloud server of the scheduling policy through the network, the smart device can obtain the amount of local execution of the base station, the task, and the amount of execution of offloading to the edge cloud server to be associated. The base station determines whether the current time slice needs to be started and the intelligent equipment to be associated.
According to the embodiment of the invention, a macro base station is arranged to integrate task information to be processed, a first network communication state, a second network communication state, first computing resource information and second computing resource information by a task engine according to received information, and construct global task information, network communication state and computing resource information; the geographical position information is delivered to an intelligent traffic engine, and the stay time of the mobile intelligent equipment in the communication range of the current macro base station is predicted by combining the intelligent traffic information; and starting a scheduling engine, and making a scheduling strategy of the current time slice by using the received information. The macro base station selects the most appropriate base station for the mobile intelligent equipment to be associated according to the current system information, and the micro base station which is not associated with the mobile intelligent equipment is enabled to enter a sleep state, so that energy is saved. Compared with the prior art, the embodiment of the invention can predict the time of the mobile intelligent device participating in the calculation through the intelligent traffic system, perform task unloading scheduling, device-base station association and base station sleep scheduling, minimize the overall energy consumption of the mobile intelligent device and the base station, simultaneously ensure the service quality requirement (mainly a delay requirement) of the application, and can be widely applied to a 5G network mobile calculation scene. Therefore, the embodiment of the invention expands the application range of the mobile edge calculation, improves the task distribution accuracy and considers the energy consumption of the system more comprehensively.
Fig. 3 is a flowchart illustrating another moving edge calculation method according to an embodiment of the present invention. Referring to fig. 3, on the basis of the foregoing embodiments, optionally, after S170, the method further includes:
and S180, providing computing service for the corresponding intelligent equipment by the edge cloud server.
And S190, after the calculation is finished, uploading the calculation result to the Internet for further analysis and processing according to the requirement, or returning the calculation result to the intelligent equipment.
The arrangement makes the moving edge calculation method more perfect.
In the above embodiments, there are various calculation methods for the scheduler engine to make the scheduling policy, and the following description specifically describes the scheduling policy, but the present invention is not limited thereto. Illustratively, embodiments of the invention introduce the following parameters to describe various variables in the framework:
si(t), the number of CPU cycles of the device in a unit time slice;
pi(si(t)), the number of CPU cycles of the device i is siPower at (t);
aij(t) marking whether the device i is associated with the base station j;
pithe upload power of device i;
rij(t), the transmission speed of the uplink of the device i and the base station j;
ri(t), the upload speed of device i;
vithe computing power of the virtual machine i at each time slice;
ci(t) marking whether to allocate computing resources for virtual machine i;
vmax(t), physical resources of the edge cloud server;
rho, the calculation density of the task;
dmaxthe latency requirements of the task;
Qi(t), local task queue of device i at time slice t;
Li(t), the device i is in a task queue of the edge cloud virtual machine;
xi(t)，Qi(t) the locally performed portion;
yi(t)，Lithe portion of (t) that is unloaded for execution;
ki(t), the amount of tasks that device i generates in time slice t;
ei(t), energy consumption of device i at time slice t;
Ej(t), energy consumption of base station j at time slice t;
bj(t), marking whether the base station j is started in the time slice t;
Pjpower consumption of base station j;
Ti(t) device i isDwell time within communication range of the front base station;
Oi(t), geographical location information of device i.
Fig. 4 is a schematic flowchart of a computing method for making a scheduling policy by a scheduling engine according to an embodiment of the present invention. Referring to fig. 4, in an embodiment of the present invention, optionally, the scheduling engine uses the received information to make a scheduling policy of the current time slice, including the following steps:
and S161, establishing a calculation formula of the local task residual quantity of the intelligent device.
And S162, establishing a task surplus calculation formula of the virtual machine in the edge cloud server in the time slice.
Wherein Q is introducedi(t) a task queue of the intelligent device i in a time slice t is represented, and the local task residual quantity of the intelligent device i is described; introduction of LiAnd (t) representing a task queue of the device i in the edge cloud virtual machine, and describing the task residual quantity of the virtual machine i in the time slice t. Let xi(t) represents QiPart of (t) executed locally, yi(t) represents QiPart of (t) offloaded to edge cloud server execution, ki(t) represents the amount of tasks newly generated by the device i in the time slice t, resulting in the queue Qi(t) and Li(t) the update rule is as follows, wherein ci(t)min(vi/ρ,Li(t)) represents the amount of tasks that virtual machine i processes at time slice t.
Qi(t+1)＝Qi(t)+ki(t)-xi(t)-yi(t)
Li(t+1)＝Li(t)+yi(t)-ci(t)min(vi/ρ,Li(t))
Introduction of Oi(T) the geographical position information of the mobile intelligent device i in the time slice T is represented in a longitude and latitude mode, the distance from a macro base station can be calculated, and the stay time T of the intelligent device i in the current base station communication range can be predicted by combining the real-time traffic state of an intelligent traffic systemi(t)。
And S163, establishing a calculation formula of the energy consumption of the intelligent equipment.
Wherein, because the edge cloud server has stable energy supply, the ITMED framework only considers the energy consumption of the smart device and the base station. Energy consumption e of the apparatusi(t) is composed of two parts of energy consumption of local calculation and communication energy consumption during task unloading, and can be expressed as follows by using a formula:
s164, establishing an energy consumption calculation formula of the base station; the base station comprises a macro base station and a plurality of micro base stations.
ITMED framework defines binary control variables bj(t) denotes sleep scheduling of the base station. When b isjWhen (t) is equal to 1, base station j is started, and when b is equal to 1jWhen (t) is 0, the base station j is turned off. Let PjRepresenting the power consumption of the base station j, the energy consumption of the base station j can be obtained as follows:
Ej(t)＝bj(t)Pj(t)。
and S165, establishing an objective function for minimizing the total energy consumption of the intelligent equipment and the base station, and limiting by constraint conditions.
The main objective of the ITMED framework is to minimize the overall energy consumption of the device and the base station and simultaneously take into account the service quality requirements (mainly delay requirements) of the application through task offload scheduling, base station sleep scheduling and device-base station association in the application scenario of the mobile intelligent device.
Wherein the constraint condition comprises at least one of the following conditions:
(1) each intelligent device establishes association with at most one base station. That is, each smart device establishes an association with at most one base station within a time slice.
Only when aijOnly if (t) is 1, does device i establish an association with base station j.
(2) Any base station can be associated with a preset number of intelligent devices at most in each time slice. That is, any base station j can associate h at most in each time slicejAn intelligent device.
(3) The association between the smart device and the base station can only be established when the uplink SINR of the smart device and the base station is greater than the target SINR. I.e. the association between the device and the base station can only be established if the uplink SINR of the device i and the base station j is greater than the target SINR.
Wherein gamma isij(t) represents the uplink SINR of device i and base station j.
(4) In any time slice, the computing resources of the virtual machine distributed by the edge cloud server for the intelligent equipment do not exceed the physical resources of the edge cloud server. The ITMED framework deploys a special virtual machine i for each intelligent device i at an edge cloud server, and the computing capacity of the virtual machine i in a unit time slice t is vi。ci(t) is a binary control variable, only if ciWhen t is 1, the edge cloud server will allocate the computing resource for the virtual machine i. Ensuring that the virtual machine computing resources distributed to the equipment by the edge cloud server do not exceed the physical resources v of the server in any time slicemax(t)。
(5) Within each time slice, the tasks that the smart device processes locally do not exceed its computational power. Within each time slice, the device does not process tasks locally beyond its computational power.
(6) In each time slice, the task amount unloaded by the intelligent equipment does not exceed the transmission capability of the intelligent equipment. I.e. the amount of tasks a device offloads does not exceed its transport capacity per time slice. y isi(t)≤ri(t)。
(7) Tasks and off-loads performed locally by the smart device within each time sliceThe sum of the executed tasks does not exceed its current task queue. I.e. the sum of the tasks performed locally by the device and the tasks performed by the off-load does not exceed its current task queue for each time slice. x is the number ofi(t)+yi(t)≤Qi(t)+ki(t)。
(8) In each time slice, the sum of the task execution time and the transmission time of the intelligent equipment does not exceed the stay time of the intelligent equipment in the communication range of the current base station. Namely, in each time slice, the sum of the task execution time and the transmission time of the equipment does not exceed the stay time of the equipment in the communication range of the current base station.
(9) Only when the base station is in the open state, the intelligent device can be associated with the base station. I.e. the device i can associate with it only if the base station j is in the on state. a isij(t)≤bij(t)。
(10) The average delay of the task can meet the delay requirement of the application. I.e. the average latency of the task can meet the latency requirements of the application.
Wherein
The method comprises the steps of calculating time delay and transmission time delay of tasks, and lambda represents the average arrival quantity of the tasks in a unit time slice of each device.
The final optimization problem (objective function) is: minimization
In summary, the embodiments of the present invention at least have the following advantages:
(1) the embodiment of the invention is closer to a real application scene, the number of the mobile intelligent devices and the number of the base stations participating in calculation in the framework are not limited, and the method has wider reference significance.
(2) The embodiment of the invention relies on a 5G network architecture and an intelligent traffic system, and the intelligent traffic system is used for accurately calculating the time of the mobile intelligent device participating in calculation, so that the calculation capability of the device can be fully utilized.
(3) The embodiment of the invention fully utilizes the sleep scheduling of the cellular base station and the association of the equipment and the cellular base station under the 5G network architecture, thereby realizing the lowest overall energy consumption of all the mobile intelligent equipment and the base station which participate in the calculation in the whole network range and being more in line with the application trend of scientific and technological development.
(4) The embodiment of the invention fully considers the energy consumption of the Internet of things equipment and the cellular base station equipment, and realizes the lowest real energy consumption.
The embodiment of the invention also provides a mobile edge calculation framework. With continued reference to FIG. 1, the moving edge calculation framework includes:
the mobile intelligent device 210 is configured to provide the macro base station with the to-be-processed task information, the first network communication state, the first computing resource information, and the geographic location information.
A micro base station 222 for providing a second network communication state to the macro base station;
an edge cloud server 230, configured to provide second computing resource information of the virtual machine to the macro base station;
the macro base station 221 is configured to, according to the received information, deliver the task information to be processed, the first network communication state, the second network communication state, the first computing resource information, and the second computing resource information to a task engine for integration, and construct global task information, network communication state, and computing resource information;
the macro base station 221 delivers the geographic position information to an intelligent traffic engine, and predicts the stay time of the mobile intelligent device in the communication range of the current macro base station by combining the intelligent traffic information;
the macro base station 221 starts a scheduling engine, and makes a scheduling strategy of the current time slice by using the received information;
the macro base station 221 notifies the smart device, the micro base station, and the edge cloud server of the scheduling policy through the network.
Optionally, the scheduling policy includes: at least one of a device task offload scheduling policy, a user-base station associated scheduling policy, and a base station sleep scheduling policy.
Optionally, the edge cloud server is further configured to provide a computing service for the corresponding smart device after the macro base station notifies the smart device, the micro base station, and the edge cloud server of the scheduling policy through the network; after the calculation is finished, the calculation result is uploaded to the internet for further analysis and processing according to the requirement, or the calculation result is returned to the intelligent equipment.
Optionally, the smart device is further configured to obtain, after the macro base station notifies the smart device, the micro base station, and the edge cloud server of the scheduling policy through the network, an amount of local execution of the base station and the task to be associated, and an amount of local execution of the task to be associated, and an amount of execution of offloading to the edge cloud server; the base station comprises a macro base station and a plurality of micro base stations.
The base station is further used for determining whether the current time slice needs to be started and the intelligent device to be associated after the macro base station informs the intelligent device, the micro base station and the edge cloud server of the scheduling strategy through the network.
Optionally, the scheduling engine is further configured to include:
establishing a calculation formula of the local task surplus of the intelligent equipment;
establishing a task surplus calculation formula of a virtual machine in a time slice in an edge cloud server;
establishing a calculation formula of the energy consumption of the intelligent equipment;
establishing an energy consumption calculation formula of the base station; the base station comprises a macro base station and a plurality of micro base stations;
an objective function is established that minimizes the overall energy consumption of the smart device and the base station, and is subject to constraints.
Optionally, the constraint condition comprises at least one of the following conditions:
each intelligent device establishes association with at most one base station;
any base station can be associated with a preset number of intelligent devices in each time slice at most;
the association between the intelligent device and the base station can be established only when the uplink SINR of the intelligent device and the base station is greater than the target SINR;
in any time slice, the computing resources of the virtual machine distributed by the edge cloud server for the intelligent equipment do not exceed the physical resources of the edge cloud server;
in each time slice, the tasks processed locally by the intelligent equipment do not exceed the computing power of the intelligent equipment;
in each time slice, the task amount unloaded by the intelligent equipment does not exceed the transmission capability of the intelligent equipment;
in each time slice, the sum of the tasks executed locally by the intelligent equipment and the tasks executed by unloading does not exceed the current task queue;
in each time slice, the sum of the task execution time and the transmission time of the intelligent equipment does not exceed the stay time of the intelligent equipment in the communication range of the current base station;
only when the base station is in an open state, the intelligent equipment can be associated with the base station;
the average delay of the task can meet the delay requirement of the application.
Optionally, the mobile edge computing framework is based on a 5G network architecture and a smart traffic system.
Optionally, the mobile smart device comprises: at least one of a smartphone and a smart car.
The mobile edge computing framework provided by the embodiment of the invention can execute the mobile edge computing method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
The embodiment of the invention also provides a computer readable storage medium. In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present invention may be executed in parallel, sequentially, or in different orders, and are not limited herein as long as the desired results of the technical solution of the present invention can be achieved.
The above-described embodiments should not be construed as limiting the scope of the invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for moving edge computation, comprising:
the mobile intelligent equipment provides task information to be processed, a first network communication state, first computing resource information and geographical position information to the macro base station;
the micro base station provides a second network communication state to the macro base station;
the edge cloud server provides second computing resource information of the virtual machine to the macro base station;
the macro base station integrates the task information to be processed, the first network communication state, the second network communication state, the first computing resource information and the second computing resource information by a task engine according to the received information, and constructs global task information, network communication state and computing resource information;
the macro base station delivers the geographic position information to an intelligent traffic engine, and the residence time of the mobile intelligent equipment in the communication range of the macro base station is predicted by combining intelligent traffic information;
the macro base station starts a scheduling engine, and a scheduling strategy of the current time slice is formulated by using the received information;
and the macro base station informs the intelligent equipment, the micro base station and the edge cloud server of the scheduling strategy through a network.
2. The method of claim 1, wherein the scheduling policy comprises: at least one of a device task offload scheduling policy, a user-base station associated scheduling policy, and a base station sleep scheduling policy.
3. The method of claim 1, wherein after the macro base station notifies the smart device, the micro base station, and the edge cloud server of the scheduling policy over a network, the method further comprises:
the edge cloud server provides computing service for the corresponding intelligent equipment;
and after the calculation is finished, uploading the calculation result to the Internet for further analysis and processing according to the requirement, or returning the calculation result to the intelligent equipment.
4. The method of claim 1, wherein after the macro base station notifies the smart device, the micro base station, and the edge cloud server of the scheduling policy over a network, further comprising:
the intelligent equipment obtains base stations to be associated, the amount of local execution of tasks and the amount of execution of unloading to the edge cloud server; the base station comprises a macro base station and a plurality of micro base stations;
the base station determines whether the current time slice needs to be started and the intelligent device to be associated.
5. The method of any of claims 1-4, wherein the scheduling engine utilizes the received information to formulate a scheduling policy for the current time slice, comprising:
establishing a calculation formula of the local task surplus of the intelligent equipment;
establishing a task surplus calculation formula of the virtual machine in the edge cloud server in the time slice;
establishing a calculation formula of the energy consumption of the intelligent equipment;
establishing an energy consumption calculation formula of the base station; the base station comprises a macro base station and a plurality of micro base stations;
an objective function is established that minimizes the overall energy consumption of the smart device and the base station, subject to constraints.
6. The method of claim 5, wherein the constraint condition comprises at least one of:
each intelligent device establishes association with at most one base station;
any base station can be associated with a preset number of intelligent devices in each time slice at most;
the intelligent device and the base station can establish association only when the uplink SINR of the intelligent device and the base station is larger than a target SINR;
in any time slice, the computing resources of the virtual machine distributed by the edge cloud server for the intelligent equipment do not exceed the physical resources of the edge cloud server;
within each time slice, the tasks processed locally by the intelligent equipment do not exceed the computing power of the intelligent equipment;
in each time slice, the task amount unloaded by the intelligent equipment does not exceed the transmission capability of the intelligent equipment;
in each time slice, the sum of the tasks executed locally by the intelligent equipment and the tasks executed by unloading does not exceed the current task queue of the intelligent equipment;
in each time slice, the sum of the task execution time and the transmission time of the intelligent equipment does not exceed the stay time of the intelligent equipment in the current base station communication range;
only when the base station is in an open state, the intelligent equipment can be associated with the base station;
the average delay of the task can meet the delay requirement of the application.
7. The method of claim 1, wherein the mobile edge computing method is based on a 5G network architecture and a smart transportation system.
8. The method of claim 1, wherein the mobile smart device comprises: at least one of a smartphone and a smart car.
9. A moving edge computing framework, comprising:
the mobile intelligent equipment is used for providing task information to be processed, a first network communication state, first computing resource information and geographical position information to the macro base station;
a micro base station for providing a second network communication state to the macro base station;
the edge cloud server is used for providing second computing resource information of the virtual machine for the macro base station;
the macro base station is used for integrating the task information to be processed, the first network communication state, the second network communication state, the first computing resource information and the second computing resource information by a task engine according to the received information to construct global task information, network communication state and computing resource information;
the macro base station delivers the geographic position information to an intelligent traffic engine, and the residence time of the mobile intelligent equipment in the communication range of the macro base station is predicted by combining intelligent traffic information;
the macro base station starts a scheduling engine, and a scheduling strategy of the current time slice is formulated by using the received information;
and the macro base station informs the intelligent equipment, the micro base station and the edge cloud server of the scheduling strategy through a network.
10. A computer-readable storage medium storing computer instructions for causing a processor to perform the moving edge calculation method of any one of claims 1-8 when executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210188436.2A CN114513838A (en) | 2022-02-28 | 2022-02-28 | Moving edge calculation method, frame, and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210188436.2A CN114513838A (en) | 2022-02-28 | 2022-02-28 | Moving edge calculation method, frame, and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114513838A true CN114513838A (en) | 2022-05-17 |
Family
ID=81553828
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210188436.2A Pending CN114513838A (en) | 2022-02-28 | 2022-02-28 | Moving edge calculation method, frame, and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114513838A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115658218A (en) * | 2022-09-09 | 2023-01-31 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for edge equipment to access cloud |
| CN117615418A (en) * | 2024-01-19 | 2024-02-27 | 北京邮电大学 | Mobile perception assisted Internet of vehicles service migration method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190058732A1 (en) * | 2015-12-01 | 2019-02-21 | Qatar Foundation for Education, Science and and Community Developmen | System and method for detection and isolation of network activity |
| CN109981753A (en) * | 2019-03-07 | 2019-07-05 | 中南大学 | A kind of system and resource allocation methods of the edge calculations of the software definition of internet of things oriented |
| WO2021194583A1 (en) * | 2020-03-23 | 2021-09-30 | Apple Inc. | Dynamic service discovery and offloading framework for edge computing based cellular network systems |
| CN113950103A (en) * | 2021-09-10 | 2022-01-18 | 西安电子科技大学 | Multi-server complete computing unloading method and system under mobile edge environment |
-
2022
- 2022-02-28 CN CN202210188436.2A patent/CN114513838A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190058732A1 (en) * | 2015-12-01 | 2019-02-21 | Qatar Foundation for Education, Science and and Community Developmen | System and method for detection and isolation of network activity |
| CN109981753A (en) * | 2019-03-07 | 2019-07-05 | 中南大学 | A kind of system and resource allocation methods of the edge calculations of the software definition of internet of things oriented |
| WO2021194583A1 (en) * | 2020-03-23 | 2021-09-30 | Apple Inc. | Dynamic service discovery and offloading framework for edge computing based cellular network systems |
| CN113950103A (en) * | 2021-09-10 | 2022-01-18 | 西安电子科技大学 | Multi-server complete computing unloading method and system under mobile edge environment |
Non-Patent Citations (1)
| Title |
|---|
| 谢玉婷: ""移动边缘计算任务卸载和基站关联协同决策问题研究"", 《计算机研究与发展》, pages 1 - 3 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115658218A (en) * | 2022-09-09 | 2023-01-31 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for edge equipment to access cloud |
| CN115658218B (en) * | 2022-09-09 | 2024-03-12 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for accessing edge equipment to cloud |
| CN117615418A (en) * | 2024-01-19 | 2024-02-27 | 北京邮电大学 | Mobile perception assisted Internet of vehicles service migration method |
| CN117615418B (en) * | 2024-01-19 | 2024-04-12 | 北京邮电大学 | Mobile perception assisted Internet of vehicles service migration method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109951821B (en) | Task unloading scheme for minimizing vehicle energy consumption based on mobile edge calculation | |
| Chen et al. | Energy-efficient task offloading and resource allocation via deep reinforcement learning for augmented reality in mobile edge networks | |
| CN111414252B (en) | Task unloading method based on deep reinforcement learning | |
| CN110647391B (en) | Edge computing method and system for satellite-ground cooperative network | |
| CN111240701B (en) | Task unloading optimization method for end-side-cloud collaborative computing | |
| CN109669768B (en) | Resource allocation and task scheduling method for edge cloud combined architecture | |
| CN109800072B (en) | Task scheduling optimization method and device based on edge calculation | |
| Li et al. | Energy-aware task offloading with deadline constraint in mobile edge computing | |
| CN112422644B (en) | Method and system for unloading computing tasks, electronic device and storage medium | |
| CN114513838A (en) | Moving edge calculation method, frame, and medium | |
| CN111586696A (en) | Resource allocation and unloading decision method based on multi-agent architecture reinforcement learning | |
| CN111355779B (en) | Service-based Internet of vehicles task unloading method and unloading device thereof | |
| CN110519776B (en) | Balanced clustering and joint resource allocation method in fog computing system | |
| CN113064671A (en) | Multi-agent-based edge cloud extensible task unloading method | |
| CN109257432A (en) | A kind of target switching method, computer installation and readable storage medium storing program for executing | |
| CN111885147A (en) | Dynamic resource pricing method in edge calculation | |
| Zhang et al. | Theoretical analysis on edge computation offloading policies for IoT devices | |
| CN114205353B (en) | Calculation unloading method based on hybrid action space reinforcement learning algorithm | |
| CN111263401A (en) | Multi-user cooperative computing unloading method based on mobile edge computing | |
| CN114363962A (en) | Collaborative edge server deployment and resource scheduling method, storage medium and system | |
| CN113645637B (en) | Method and device for unloading tasks of ultra-dense network, computer equipment and storage medium | |
| CN114595049A (en) | Cloud-edge cooperative task scheduling method and device | |
| CN111988787B (en) | Task network access and service placement position selection method and system | |
| CN115408072A (en) | Rapid adaptation model construction method based on deep reinforcement learning and related device | |
| Tian et al. | Asynchronous federated learning empowered computation offloading in collaborative vehicular networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |